Op amp inductor-based wah

[effdub]

I was reading through RG Keen's article on how wah's work:
http://www.geofex.com/Article_Folders/wahpedl/wahped.htm

And I saw his barebones schematic for an opamp + inductor wah and thought it would be a fun project to work on in concert. So here is the original:



Could be a good starting point for all sorts of fun funky stuff. Or maybe even just a wah. 


I'm having a bit of trouble sorting out what the resistors values should be. Some of them I can probably guess reasonably well, but I don't get what all of them are doing.

I'll start breadboarding on Saturday and see what I can come up with by guessing. Or maybe someone could suggest a mathematical (gasp!) approach to help me make educated guesses? That'd be cool.

Also, why is the non-inverting input of the "lower" op amp tied to ground? It looks to be set up as an inverting amplifier, but it seems that you wouldn't tie + input to ground with a power supply with only one polarity. Shouldn't it be tied to VREF?

[earthtonesaudio]

The little triangles represent the midpoint of the supply/supplies.  So if using a single supply, they would be V+/2, and if using a bipolar supply, then they're ground.

[edit]Added some suggested values.  I got most of these from the "Basic Wah Circuit" schem at GEO.  Note that the only values that are transplanted directly are the things between the input of the non-inverting stage and the output of the inverting stage; the gain-setting resistors on the op-amps are somewhat arbitrary, as is the wah pot value.  I just made the wah pot 1/10th the input impedance of the inverting stage to minimize loading.

[effdub]

Hooray, I guess a couple of the values the same as you!

The little triangles represent the midpoint of the supply/supplies.  So if using a single supply, they would be V+/2, and if using a bipolar supply, then they're ground.

So nothing in the circuit is actually really grounded? That's interesting.

[earthtonesaudio]

Well, they should be "AC grounded" which means that Vref should be a low impedance voltage source.  Probably it would be fine to use two equal value resistors and slap a large cap to ground from their junction, like you see with most single supply op-amp circuits.  A buffered Vref would be better of course, but maybe overkill.

The schematic also implies that the input signal is referenced to/riding on this same voltage.  If the input is referenced to some other voltage, you should include a DC blocking cap in series with the input resistor.


It's worth mentioning that there is actually an "AC ground" in the discrete wah circuit as well.  The 4.7uF cap on one end of the inductor is a low impedance for signal currents and its voltage remains more or less constant.  It's not as obvious as the "two resistors, one capacitor" reference you might be used to, but it serves the same purpose.

[effdub]

Would there be any advantage in a wah circuit to have a bi-polar supply?


Also, why would anyone bother with a filtered voltage divider if they could just use a cap to ground?

[earthtonesaudio]

If I happened to have a bipolar supply handy I would use it; but otherwise I don't see an advantage.

The cap to ground is the filtering for the voltage divider.  It stores up charge so when the signal needs some, it can take it from the cap.  If the signal takes charge from the cap (but not the resistors) the voltage stays constant.

[effdub]

The cap to ground is the filtering for the voltage divider. 

Duh, yeah. I just looked at the schematic again (for the BJT wah) and now it makes sense.